Page 1 of 1

Track Percentage of times a player goes first

PostPosted: Mon Jun 09, 2014 7:18 pm
by universalchiro
Concise description:
  • In the world of how many times we roll 1s, 2s, 3s, 4s, 5s, 6s, and how many kills we have versus losses we have versus a particular player, there is still one aspect of the game that is not tracked, not monitored and it should be. And that is how many times one player has gone first versus another player, or how many times one team has gone first versus another team. Sure a game can be decided by who has more 6s and who has more 2s, but discounting that for it's part of the game, a game can be decided by who goes first. How many times have you played a game that was decided by who goes first? We have all played a game where the player going first has a decided advantage. Especially in 1v1 and team games. (Maybe not so much in multiple player standard games) I noticed a pattern when playing high quality players in Conquer Rome and let's say the rolls are even, then the team going first still has an advantage. I always feel like I'm playing catch up when I go second. My thoughts tend to drift to, "I have to deal with their bonus and pick up my own bonus". In a tournament I played, I picked up on a trend that if my team went first, the game was usually an easy victory and when my team went second, it was a fight to contend for victory. Unfortunately, and the reason for this suggestion, I counted the number of starts versus other teams and my team had a couple times where we went second in both games versus one team, which adversely affected the game results. So it is a step in the right direction to reveal the ratio, the percentages of who goes first, who goes second in 1v1 and team games.

    However, this topic/suggestion is multifaceted, it doesn't deal with just one aspect of going first or second, but two distinct aspects of going first or going second. Conquer Club keeps track of our rolls and Battle Outcomes and adding a tab that keeps track of going first that lead to victory, going first that lead to loss, going second that lead to victory and going second that lead to loss, for 1v1 and team games is important information of revealing how well a player/team performs while going second in the disadvantageous position and well a player/team performs when they have the advantage of going first.

    Another aspect of this topic is adjusting the scoring to reflect the advantage of going first and discount the disadvantage of going second.

Specifics/Details:
    1. To shed light on who goes first in 1v1's and team games. To keep a record of how many times a player goes first or second in 1v1 and team games.
    a) Keeps ratio of how a player takes care of business with the advantage of going first.
    b) Keeps ratio of how a player battles back from the disadvantage position of going second.

    2. To give more weight to victory while going second by rounding up on fractional points, and less weight to victory when going first by rounding down fractional points and conversely one gets more weight for a victory going second by rounding up fractional point and one gets less points taken away by a loss going second with rounding fractional points down.
    a) If a player wins going first and the points due are 20.79, then round down to 20.
    b) If a player wins going second and the points due are 20.10, then round up to 21.
    c) If a player losses going first and the points due are -20.49 then round up to -21.
    d) If a player losses going second and the points due are -20.70, then round down to 20.

How this will benefit the site and/or other comments:
  • To have another tab for statistics for a gamer? This is a win-win, all gamers love statistics. players will want to know how they take care of business with the advantage position of going first, and players want to know how they battle back in the disadvantageous position of going second. Also, this will lead to ratios revealing how often one player gets to go first or second and will lead to tournaments that implement alternating going first to start the game in the distant future. Which will eventually lead to more fair game, which leads to a more enjoyable game, which leads to increased player retention. In summary, this portion of tracking who goes first or second will lead to more fair play in the future, but to get there, we need this tracking statistics first.

    As far as handicapping fractional points for going first or second, with one game this doesn't add up to much, but with 10,000s games under a players belt or 15,000 games per day for the site, this discounting points, even though it's fractional, will mount to a great deal of points being more fairly distributed for victories that occurred in the tougher position of going second in 1v1 and team games. There are almost always fractional points rounded up or down after each game, so this will see results with each game. In addition, players that feel an unfair advantage in a tournament when they play a team twice and both times they went second, which adversely affects the game and adversely affects the enjoyment of the game, will have some sense that the handicapping of fractional points will ease the pain of losing points from going second.

Re: Track Percentage of times a player goes first

PostPosted: Mon Jun 09, 2014 11:16 pm
by macbone
Sure, the team that goes first has an inherent advantage. They might start with a bonus, or are able to break another team's bonus, or have the opportunity to take and secure a bonus, or reduce the next opponent's number of terts, or any of a number of other advantages. After all, the dice odds favor attacking, and if a team gets a chance to either attack first or set up an even bigger assault by building a round or two undisturbed, they'll benefit from it. The only counter to this is the dice.

However, I definitely wouldn't adjust the score for a team that goes second, third, fourth, etc. That's simple luck. We don't adjust the score for players who have lower than average dice in a game, or who have a poorer drop, or who have a colonel on the team who is nevertheless clueless.

Look at a game taken even more seriously, chess. Chess sites track the percentage of times a player plays with black or white and the rate of wins for each color. Playing white is an inherent advantage, and a player has a higher chance of winning the game when playing white, but there is no adjustment in the score in this instance, not on chess sites, a more casual arena, or in competitions, where the stakes can be much higher.

Tracking the number of times a player goes 1st, 2nd, 3rd, 4th, 5th, 6th, etc. might be interesting, but that's a ton of stats. Even if it's limited to team games, we have match-ups like 3-team quads and 6-team doubs play available right now. Or are you solely concerned with 1v1 and two-team games? If so, will the system provide statistics only for those games and ignore games with 3 or more players/teams?

It might be an interesting project for an add-on.

Re: Track Percentage of times a player goes first

PostPosted: Wed Jun 11, 2014 1:35 pm
by Kaskavel
macbone wrote:
Look at a game taken even more seriously, chess. Chess sites track the percentage of times a player plays with black or white and the rate of wins for each color. Playing white is an inherent advantage, and a player has a higher chance of winning the game when playing white, but there is no adjustment in the score in this instance, not on chess sites, a more casual arena, or in competitions, where the stakes can be much higher.




This is not entirely correct. In chess white player scores an average 54%. This varies from 58% when both opponents are top grandmasters (true, those games have an increased chance to get drawn, but black wins quite rarely) to almost 50% when both players are amateurs (and thus the game hardly gets decided by who went first)
But, disadvantage of playing black does get recognized sometimes. Many tournaments decide the winner of a tournament in case of a opint tie based upon "most blacks played" or "most blacks won" and such stuff. There were even existed some propositions that the draw offers 0.55 points to black and 0.45 to white, instead of 0.5-0,5.
Anyway, I think the OPP talks too much for something too simple. He is right. The first part of the suggestion is correct and simple. I vote for it. And I also want to add an argument for it. It will also lead to map comparisons, showing which maps favor much going first. For example, in games between 2 generals, red may score 99% in City Mongul, 55% in Feudal, 75% in classic etc.
Second part of the suggestion finds me also voting for it, but I realize it will not be implemented...People will just react, so I would suggest to him, not to press the matter further. If first part is implemented, and people will start watching the stats, they will eventualy comprehend why an adjustment in points won/lost is natural.
Finaly, just because we cannot adjust points for good/bad dice, is not an argument against doing the same thing for something that we can control.

Re: Track Percentage of times a player goes first

PostPosted: Thu Jun 12, 2014 12:53 am
by macbone
Yes, Kaskavel, that's true. The World Chess Federation does include the number of wins with black as one of the methods of determining a tie-breaker, generally after comparing the head-to-head wins and most wins: http://www.fide.com/fide/handbook.html? ... w=category. Alternate scoring methods beyond the 1 pt for a win, 1/2 point for a draw, and 0 pts for a loss are generally the exception, though, not the rule.

However, tournaments do try to balance the number of games played with each color. In general, players alternate colors each round whenever possible.

Re: Track Percentage of times a player goes first

PostPosted: Thu Jun 12, 2014 2:37 pm
by Kaskavel
Yes, but all open tournaments have an even number of rounds, typically 7 or 9, for reasons that get too irrelevant for this forum to analyze. Anyway, let's not spoil the OPP suggestion...